转基因油菜的基因流及生态风险

综合评述了转基因油菜的基因流及其生态风险.油菜作为最早的转基因作物之一目前已在加拿大和澳大利亚大面积商业化应用.(常)异花授粉作物油菜的天然异交率可达30%左右,也易与其它芸苔属作物杂交,因此转基因油菜的生态风险已引起各国科学家的高度重视.转基因油菜主要通过与其野生近缘种的花粉交换和与非转基因油菜的花粉交换两种方式进行花粉的输出.基因可能逃逸到相关野生近缘种,但在大田环境下能够得到杂种的可能性很小;由于基因的漂流在油菜田块间确实存在,因此在种植转基因油菜的过程中必须考虑其间隔距离.     

Inheritance of rapeseed (Brassica napus)-specific RAPD markers and a transgene in the cross B. juncea x (B. juncea x B. napus)

We have examined the inheritance of 20 rapeseed (Brassica napus)-specific RAPD (randomly amplified polymorphic DNA) markers from transgenic, herbicide-tolerant rapeseed in 54 plants of the BC₁ generation from the cross B. junceax(B. junceaxB. napus). Hybridization between B. juncea and B. napus, with B. juncea as the female parent, was successful both in controlled crosses and spontaneously in the field. The controlled backcrossing of selected hybrids to B. juncea, again with B. juncea as the female parent, also resulted in many seeds. The BC₁ plants contained from 0 to 20 of the rapeseed RAPD markers, and the frequency of inheritance of individual RAPD markers ranged from 19% to 93%. The transgene was found in 52% of the plants analyzed. Five synteny groups of RAPD markers were identified. In the hybrids pollen fertility was 0–28%. The hybrids with the highest pollen fertility were selected as male parents for backcrossing, and pollen fertility in the BC₁ plants was improved (24–90%) compared to that of the hybrids.     

Polymorphism for interspecific hybridisation within a population of wild radish (Raphanus raphanistrum) pollinated by oilseed rape (Brassica napus)

The within-population polymorphism of wild radish (Raphanus raphanistrum) for interspecific hybridisation with two cultivars of oilseed rape (Brassica napus) was investigated by hand crossing experiments and fluorescence microscopy. Wide variability among plants was observed in the ability of oilseed rape pollen to germinate on the wild radish stigma; the frequency of pistils showing pollen tubes ranged from 0 to 1, depending on the female plant. The ratio of fertilised ovules to the total number of ovules in ovaries where pollen tubes arrived ranged from 0.02 to 0.51. Overall, the results provide evidence for the presence of different phenotypes. In 40% of the plants, pistils had no or very few pollen tubes and few fertilised ovules. In 23%, the foreign pollen tubes grew through the style towards the ovary, but had low ovule fertilisation efficiency. The remaining 37% showed a large number of pollen tubes in the style and frequent ovule fertilisation, and two plants showed no difference between foreign and conspecific pollen. With regard to post-zygotic barriers, pollen germination and ovule fertilisation represent minor barriers to interspecific hybridisation between oilseed rape and wild radish. It is suggested that the effectiveness of these barriers could be improved through plant breeding; this could reduce the risk of gene flow from transgenic oilseed rape to populations of wild relatives. Received: 15 April 2001 / Accepted: 24 May 2001     

Fitness of backcross six of hybrids between transgenic oilseed rape (Brassica napus) and wild radish (Raphanus raphanistrum)

The process of introgression between a transgenic crop modified for better agronomic characters and a wild relative could lead potentially to increased weediness and adaptation to the environment of the wild species. However, the formation of hybrid and hybrid progeny could be associated with functional imbalance and low fitness, which reduces the risk of gene escape and establishment of the wild species in the field. Our work compares the fitness components of parents and different types of backcross in the sixth generation of hybrids between transgenic oilseed rape (Brassica napus, AACC, 2n = 38) resistant to the herbicide glufosinate and wild radish (Raphanus raphanistrum, RrRr, 2n = 18). The backcross with oilseed rape cytoplasm (OBC) has a fitness value 100 times lower than that of the backcross with wild radish cytoplasm (RBC). The herbicide-resistant RBC has similar growth to the susceptible RBC, but final male and female fitness values are two times lower. In turn, susceptible RBC exhibit similar fitness to the control wild radishes. The relative fitnesses of the different types are the same whether or not they grow under competitive conditions. The consequence on fitness of the chromosome location of the transgene conferring resistance and the relevance of these results to the impact of gene flow on the environment are discussed.     

Opportunities for gene transfer from transgenic oilseed rape (Brassica napus) to related species

Before novel transgenic plant genotypes are grown outside containment facilities and evaluated under field conditions, it is necessary to complete a risk assessment to consider the possible consequences of that release. An important aspect of risk assessment is to consider the likelihood and consequences of the transgene being transferred by cross-pollination to related species, including other crops, weeds and ruderal populations. The purpose of this report is to review the literature to assess the ease with whichBrassica napus can hybridize with related species. The evidence for hybridization is considered at three levels: a) by open pollination, b) by hand pollination and c) by the use ofin vitro ovule and embryo rescue techniques; and also examines the fertility and vigour of the F₁, F₂ and backcross generations. Four species are reported to hybridize withB. napus by open pollination:B. rapa andB. juncea using fully fertile parents; andB. adpressa andR. raphanistrum using a male-sterileB. napus parent. Seventeen species are reported to form hybrids (including the four species above) withB. napus when pollination is carried out manually. At least 12 of these species were unable to form F₂ progeny, and eight were unable to produce progeny when the F₁ was backcrossed to one of the parental species. Many factors will influence the success of hybridization under field conditions, including: distance between the parents, synchrony of flowering, method of pollen spread, specific parental genotypes used, direction of the cross and the environmental conditions. Even where there is a possibility of hybridization betweenB. napus and a related species growing in the vicinity of a release, poor vigour and high sterility in the hybrids will generally mean that hybrids and their progeny will not survive in either an agricultural or natural habitat.     

Association of RFLP markers and biomass heterosis in trigenomic hybrids of oilseed rape ( Brassica napus x B. campestris)

Laboratory screening with DNA-based markers and field measurements of biomass production were carried out on each of the 120 trigenomic hybrids, obtained by interspecific hybridization between Brassica napus (AACC) and Brassica campestris (A'A'). The objective of this study was to elucidate the relationship between molecular markers and biomass heterosis of the interspecific hybrid between B. napus and B. campestris, which has been explored practically in rapeseed production for many years. The experiment was first carried out on 65 trigenomic hybrids in 1999. The average over-mid-parent heterosis of biomass production was around 30%, and the highest value was 175.4%. In the following year, the observation was expanded to 120 trigenomic hybrids and the best average over-mid-parent heterosis was 93%. A total of 1,477 DNA fragments, generated by Southern hybridization with 50 Brassica cDNA clones and 25 Arabidopsis EST clones, was scored across their parental lines. One hundred and twenty six and 215 fragments were identified as significantly associated with biomass production respectively in the 2 successive years. Using these active markers, a statistical model to resolve the heterosis is proposed and a new way to make use of the subgenomic heterosis is also discussed.     

Spontaneous hybridization between oilseed rape (Brassica napus) and weedy B. campestris (Brassicaceae): a risk of growing genetically modified oilseed rape

Frequencies of spontaneous hybridization between oilseed rape (Brassica napus L.) and weedy Brassica campesíris ssp. campestris L. were measured in agricultural fields. Hybrids were identified by enzyme electrophoresis, random amplified polymorphic DNA (RAPD) analysis, chromosome counting, morphology, and pollen fertility. When the two species were mixed 1:1, B. campestris produced 13% hybrid seeds and oilseed rape 9%. In two experiments with single plants of the self-incompatible B. campestris widely spaced within fields of oilseed rape, 56% and 93% hybrid seeds were produced. Analysis of a weedy population of B. campestris in oilseed rape revealed 60% hybrid seeds. Backcrossing of the hybrids to the weedy species seems to occur as supported by the finding in a natural population of two B. campestris-like plants with a marker specific to oilseed rape. The results suggest that transgenes could be dispersed from oilseed rape to B. campestris.     

Molecular markers for low-glucosinolate alleles in oilseed rape (Brassica napus L.)

Two recent studies have mapped QTLs associated with the level of seed glucosinolates in oilseed rape (Brassica napus L.). It was likely that the two most significant QTLs identified in each study were the same, as they were linked to RFLP alleles identified by common DNA probes. To investigate the utility of these probes in breeding programmes, they were used to study RFLPs in a range of low- and high-glucosinolate cultivars and breeding lines. It was shown that all low glucosinolate spring and winter cultivars possessed a specific RFLP fragment identified by probe wg3f7 which is linked to theGSL-1 QTL, and all high-glucosinolate cultivars possessed a specific RFLP fragment identified by probe wg7a8, which is linked to theGSL-2 QTL. Cultivar Ariana, which has intermediate levels of glucosinolates possessed the low-glucosinolate fragment atGSL-1 but the high-glucosinolate fragment atGSL-2. A similar result was found with the cvs. Martina and Bronowski which have intermediate and variable levels of glucosinolates. There were no other RFLP fragments identified by other DNA probes which were specific to either the low- or high-glucosinolate phenotypes. The use of probes wg3f7 and wg7a8 in selection of low-glucosinolate lines in breeding programmes is discussed.     

Construction of an oilseed rape (Brassica napus L.) genetic map with SSR markers

We constructed a Brassica napus genetic map with 240 simple sequence repeats (SSR) primer pairs from private and public origins. SSR, or microsatellites, are highly polymorphic and efficient markers for the analysis of plant genomes. Our selection of primer pairs corresponded to 305 genetic loci that we were able to map. In addition, we also used 52 sequence-characterized amplified region primer pairs corresponding to 58 loci that were developed in our lab. Genotyping was performed on six F2 populations, corresponding to a total of 574 F2 individual plants, obtained according to an unbalanced diallel cross design involving six parental lines. The resulting consensus map presented 19 linkage groups ranging from 46.2 to 276.5 cM, which we were able to name after the B. napus map available at , thus enabling the identification of the A genome linkage groups originating from the B. rapa ancestor and the C genome linkage groups originating from the B. oleracea ancestor in the amphidiploid genome of B. napus. Some homoeologous regions were identified between the A and the C genomes. This map could be used to identify more markers, which would eventually be linked to genes controlling important agronomic characters in rapeseed. Furthermore, considering the good genome coverage we obtained, together with an observed homogenous distribution of the loci across the genome, this map is a powerful tool to be used in marker-assisted breeding. Electronic Supplementary Material Supplementary material is available for this article at     

Persistence of oilseed rape (Brassica napus L.) outside of cultivated fields

It is widely assumed that most cultivated plants cannot persist in natural or semi-natural habitats. Most people thus assume that the plants growing outside of fields (in particular oilseed rape along roadsides) find their origins in the current or previous year’s cultivation of that crop. One consequence of this assumption is that the identity of plants growing on road verges is thought to reflect one of the cultivars currently or recently cultivated, while another consequence is the widespread belief that transgenic plants can be simply managed and controlled by stopping their cultivation. Our work shows that this assumption is false. We identify relict plants of a now unmarketable cultivar type of oilseed rape which have persisted in a semi-natural habitat (road verges) for at least 8 years according to farmer surveys in the studied area. More generally, we confirm that the dynamics of the feral oilseed rape plants of road verges is more complex than those resulting from spillage from agricultural machines or from neighbouring arable fields cultivated the previous year. Within the scope of transgenic oilseed rape cultivation, these results suggest that more studies on the dynamics of feral oilseed rape are needed in order to assess more precisely the risks of its invasiveness and its potential impact on genetic pollution between GM fields and non-GM fields. Received: 15 June 2000 / Accepted 31 July 2000     

Gene flow from wheat (Triticum aestivum L.) to jointed goatgrass (Aegilops cylindrica Host.), as revealed by RAPD and microsatellite markers

In order to estimate the potential of gene flow between wheat (Triticum æstivum L.) and jointed goatgrass (Aegilops cylindrica Host.), we carried out mixed pollinations in experimental and natural conditions. A set of species-specific RAPD (random amplified polymorphic DNA) and microsatellite markers were used to detect the presence of parental markers in the progeny of the plants used in these experiments. No hybrids were found within the offsprings of the plants used for the greenhouse experiments, while 85 Ae. cylindrica×T. æstivum hybrids were found within 2400 analyzed F₁ plants resulting from the field pollinations. The hybridization rates for individuals of different populations of the wild species differed considerably: 1% for two populations known for more than 90 years versus 7% for a newly discovered population. Most of the hybrids were completely sterile, but five of them produced 13 seeds (BC₁) by backcross with Ae. cylindrica. Twelve seeds germinated and generated viable and partly fertile plants. About 25% of the wheat specific RAPD markers were found in the BC₁ plants, indicating that introgression of wheat DNA into Ae. cylindrica is possible. In addition, one microsatellite marker, known to be situated on the D genome (a genome shared by both species), was also found in the BC₁ plants.     

Association of gene-linked SSR markers to seed glucosinolate content in oilseed rape (Brassica napus ssp. napus)

Breeding of oilseed rape (Brassica napus ssp. napus) has evoked a strong bottleneck selection towards double-low (00) seed quality with zero erucic acid and low seed glucosinolate content. The resulting reduction of genetic variability in elite 00-quality oilseed rape is particularly relevant with regard to the development of genetically diverse heterotic pools for hybrid breeding. In contrast, B. napus genotypes containing high levels of erucic acid and seed glucosinolates (++ quality) represent a comparatively genetically divergent source of germplasm. Seed glucosinolate content is a complex quantitative trait, however, meaning that the introgression of novel germplasm from this gene pool requires recurrent backcrossing to avoid linkage drag for high glucosinolate content. Molecular markers for key low-glucosinolate alleles could potentially improve the selection process. The aim of this study was to identify potentially gene-linked markers for important seed glucosinolate loci via structure-based allele-trait association studies in genetically diverse B. napus genotypes. The analyses included a set of new simple-sequence repeat (SSR) markers whose orthologs in Arabidopsis thaliana are physically closely linked to promising candidate genes for glucosinolate biosynthesis. We found evidence that four genes involved in the biosynthesis of indole, aliphatic and aromatic glucosinolates might be associated with known quantitative trait loci for total seed glucosinolate content in B. napus. Markers linked to homoeologous loci of these genes in the paleopolyploid B. napus genome were found to be associated with a significant effect on the seed glucosinolate content. This example shows the potential of Arabidopsis-Brassica comparative genome analysis for synteny-based identification of gene-linked SSR markers that can potentially be used in marker-assisted selection for an important trait in oilseed rape. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Barriers to gene flow from oilseed rape (Brassica napus) into populations of Sinapis arvensis

One concern over growing herbicide-tolerant crops is that herbicide-tolerance genes may be transferred into the weeds they are designed to control. Brassica napus (oilseed rape) has a number of wild relatives that cause weed problems and the most widespread of these is Sinapis arvensis (charlock). Sinapis arvensis seed was collected from 102 populations across the UK, within and outside B. napus-growing areas. These populations were tested for sexual compatibility with B. napus and it was found that none of them hybridized readily in the glasshouse. In contrast to previous studies, we have found that hybrids can be formed naturally with S. arvensis as the maternal parent. Six diverse B. napus cultivars (Capricorn, Drakkar, Falcon, Galaxy, Hobson and Regent) were tested for their compatibility with S. arvensis but no cultivar hybridized readily in the glasshouse. We were unable to detect gene transfer from B. napus to S. arvensis in the field, confirming the extremely low probability of hybridization predicted from the glasshouse work.     

Genetic manipulation in cultivars of oilseed rape (Brassica napus) using Agrobacterium

Summary The response of oilseed rape cultivars to infection with Agrobacterium tumefaciens and A. rhizogenes and the possibility of regenerating genetically transformed oilseed rape plants were examined. The frequency at which Agrobacterium induced galls or hairy-roots on in vitro cultured plants ranged from 10% to 70%, depending on the cultivar. From galls induced by the tumorigenic strain T37, known to be strongly shoot inducing on tobacco, roots developed frequently. Occasionally, shoots formed and some of these produced tumour cell specific nopaline. Attempts to grow the transformed shoots into plants have so far been unsuccessful. Whole plants transformed with Ri-T-DNA, however, were regenerated. These had crinkled leaves and abundant, frequently branching roots that showed reduced geotropism, similar to previously isolated Ri T-DNA transformed tobacco and potato plants. The transformed oilseed rape plants flowered, but failed to form seeds.     

Sources of genetic diversity in feral oilseed rape (Brassica napus) populations

Feral oilseed rape populations on roadsides and along other semi-natural habitats contribute to the uncertainty regarding the potential risk of transgene escape from genetically modified (GM) oilseed rape to wild or weedy populations. Knowledge about the underlying mechanism of establishment, persistence, and the evolutionary potential of feral populations is a prerequisite for a successful risk assessment and of high policy relevance. We analysed the origin, persistence, and genetic variation of feral oilseed rape populations in northwest Germany based on a field survey (2004–2007) and nuclear microsatellite markers. A maximum-likelihood approach was used to allocate individuals of feral populations to their source varieties and to hybrids between these varieties. The majority (72%) of the analysed sites was inhabited by oilseed rape for at least 2 years. The proportion of feral populations setting seeds varied between years (30–48%) and was higher than in comparable studies. Within-population genetic diversity of feral oilseed rape populations was higher than that of the common grown varieties studied. Repeated escapes of different varieties and hybridisation between these varieties were identified as the most important sources of genetic variation in feral oilseed rape populations. Up to four different source varieties were identified in feral populations. The detection of three hybrids between different varieties indicates that feral oilseed rape populations may persist via self-recruitment. Our results highlight the evolutionary potential of feral oilseed rape populations and have consequences for approaches to regulate the coexistence of genetically modified and non-GM oilseed rape varieties in the same region. Feral populations may serve as stepping stones for intraspecific and interspecific gene flow and can promote persistence of transgenes outside cultivation. The likelihood of introgressive interspecific hybridisation depends on the abundance of crossing partners, which may vary strongly from region to region.     

Hybridisation between oilseed rape (Brassica napus) and tetraploid Brassica rapa under field conditions

Cross-compatible relatives of crop species contribute to the uncertainty regarding the potential risk of transgene escape from genetically modified varieties. The most successful crossing partner of oilseed rape (Brassica napus L.) is diploid Brassica rapa L. Variation of ploidy level among B. rapa cultivars has, until recently, been neglected in the context of gene flow and hybridisation with oilseed rape. We estimated the extent of hybridisation between autotetraploid B. rapa varieties (female) and B. napus (pollen donor) under experimental field conditions. Morphology, variation of relative DNA amount, and microsatellite markers were used to distinguish between intraspecific offspring of tetraploid B. rapa and interspecific hybrids with B. napus. Of 517 seed progenies of tetraploid B. rapa, 45 juvenile plants showed species specific morphological traits of oilseed rape. The detection of putative hybrids based on variation in relative DNA amounts was problematic due to the occurrence of aneuploidy. In total, 84 offspring showed relative DNA amounts deviating from tetraploid B. rapa, four of which were hexaploids. Of the 205 offspring analysed at three microsatellite loci, 67 had oilseed rape alleles. Based on molecular evidence a minimum hybridisation rate of 13.0% was estimated. A few mother plants accounted for the majority of hybrids. The mean pollen viability of hybrids between B. napus and tetraploid B. rapa (80.6%) was high in comparison with mean pollen viability of triploid hybrids between B. napus and diploid B. rapa. Therefore, the occurrence of tetraploid B. rapa should be taken into consideration when estimating the likelihood of gene flow from oilseed rape to close relatives at the landscape level. Tetraploid B. rapa is a common component of several seed mixtures and establishes feral populations in northwest Germany. Assuming a similar abundance of diploid and tetraploid B. rapa, gene flow from B. napus to tetraploid may be more likely than gene flow to diploid B. rapa.     

Preferential exclusion of hybrids in mixed pollinations between oilseed rape (Brassica napus) and weedy B. campestris (Brassicaceae)

In most experimental hybridizations between oilseed rape (Brassica napus) and weedy B. campestris, either intra- or interspecific pollen has been applied to individual flowers. Under field conditions, however, stigmas will often receive a mixture of the two types of pollen, thereby allowing for competition between male gametophytes and/or seeds within pods. To test whether competition influences the success of hybridization, pollen from the two species was mixed in different proportions and applied to stigmas of both species. The resulting seeds were scored for paternity by isozyme and randomly amplified polymorphic DNA analysis. Using data on the proportion of fully developed seeds and the proportion of these seeds that were hybrids, a statistical model was constructed to estimate the fitness of conspecific and heterospecific pollen and the survival of conspecific and heterospecific zygotes to seeds. B. campestris pollen in B. napus styles had a significantly lower fitness than the conspecific pollen, whereas no difference between pollen types was found in B. campestris styles. Hybrid zygotes survived to significantly lower proportions than conspecific zygotes in both species, with the lowest survival of hybrid zygotes in B. napus pods. This is in contrast to the higher survival of hybrid seeds in B. napus than in B. campestris pods when pollinations are made with pure pollen. Altogether, the likelihood of a foreign pollen grain producing a seed was much lower on B. napus than on B. campestris. In addition, pods on B. napus developed to a lower extent the more heterospecific pollen was in the mix, whereas this had no effect on B. campestris.     

Modelling and estimating pollen movement in oilseed rape (Brassica napus) at the landscape scale using genetic markers

Understanding patterns of pollen movement at the landscape scale is important for establishing management rules following the release of genetically modified (GM) crops. We use here a mating model adapted to cultivated species to estimate dispersal kernels from the genotypes of the progenies of male-sterile plants positioned at different sampling sites within a 10 x 10-km oilseed rape production area. Half of the pollen clouds sampled by the male-sterile plants originated from uncharacterized pollen sources that could consist of both large volunteer and feral populations, and fields within and outside the study area. The geometric dispersal kernel was the most appropriate to predict pollen movement in the study area. It predicted a much larger proportion of long-distance pollination than previously fitted dispersal kernels. This best-fitting mating model underestimated the level of differentiation among pollen clouds but could predict its spatial structure. The estimation method was validated on simulated genotypic data, and proved to provide good estimates of both the shape of the dispersal kernel and the rate and composition of pollen issued from uncharacterized pollen sources. The best dispersal kernel fitted here, the geometric kernel, should now be integrated into models that aim at predicting gene flow at the landscape level, in particular between GM and non-GM crops.     

Low probability of chloroplast movement from oilseed rape (Brassica napus) into wild Brassica rapa

Pollen-mediated movement of transgenes from transplastomic oilseed rape (Brassica napus) into wild relatives will be avoided if chloroplasts are maternally transmitted. We assess the probability of chloroplast exchange between conventional oilseed rape and wild Brassica rapa to model the future behavior of transplastomic cultivars. Primers specific to cpDNA were used to demonstrate maternal inheritance of chloroplasts in 47 natural hybrids between cultivated B. napus and wild B. rapa. We conclude that there will be no or negligible pollen-mediated chloroplast dispersal from oilseed rape. Transgene introgression could still occur in mixed populations, however, if B. napus acted as the recurrent female parent. Rate of transfer would then depend on the abundance of mixed populations, their persistence as mixtures, and hybridization frequency within stands. A low incidence of sympatry (0.6-0.7%) between wild B. rapa and cultivated B. napus along the river Thames, UK, in 1997 and 1998, suggests mixed stands will form only rarely. Eighteen feral populations of B. napus also showed a strong tendency toward rapid decline in plant number, seed return, and ultimately, extinction within 3 years. Conversely, hybrid production is significant in mixed stands, and the absence of control practices means that oilseed rape will have slightly greater persistence. We infer that some introgression from transplastomic B. napus into B. rapa is inevitable in mixed populations even though such populations will occur infrequently and will tend to lose B. napus plants relatively quickly. Chloroplast exchange will be extremely rare and scattered.